Factors leading to falls in transfemoral prosthesis users: a case series of prosthesis-side stumble recovery responses.

The study included only six transfemoral prosthesis users, so it cannot describe how most prosthesis users respond. All participants had K3 or greater function, so the results may not apply to lower-function users. A fall was defined as more than 50% body-weight support in the harness, not a true fall to the ground. The overhead harness, treadmill, goggles, noise, and counting task may have changed how participants moved. The treadmill speed was fixed at 0.8 m/s and may not have matched each person's preferred speed. The control group was measured at 1.1 m/s, not the same speed as the prosthesis users. The study was not designed to compare C-Leg users with users of the other two knees, and the paper says device response may not be isolated from the participant response. The proposed stance support, swing extension, flexion initiation, impedance changes, strength, and perturbation training were not tested.

The paper names National Institutes of Health as the funder. The supplied text does not report other conflicts of interest.

Do not read the 54% fall rate as a real-world fall probability. The trials were on a treadmill, the participants were distracted, and a fall was recorded when the harness took more than 50% body-weight support. Do not read the proposed device changes or training ideas as tested interventions.